ABSTRACT NUMBER: ESOC2026A1241 NURSE-LED TRIAGE OF PRE-HOSPITAL STROKE CODES REDUCES EASILY RECOGNIZED MIMICS WITHOUT AFFECTING STROKE YIELD

Abstract Background and aims Pre-hospital stroke code activation by emergency medical services (EMS) is essential for timely treatment in the emergency department (ED) but is frequently triggered by non-stroke conditions (“stroke mimics”). A protocol-driven pilot program was implemented allowing ED nurses to cancel pre-hospital activated stroke codes when predefined criteria were met for non-stroke presentations or acute treatment ineligibility. We evaluated the impact of this pilot on the distribution of final diagnoses and stroke mimics. Methods We retrospectively compared consecutive ED stroke code activations during two periods: pre-pilot (n = 258) and post-pilot (n = 493). Final diagnoses were classified as confirmed stroke or stroke mimic, and mimics were further categorized (e.g., infection, dizziness, seizure, functional). Proportions of stroke and mimic subtypes were compared between periods. Results The proportion of confirmed strokes remained stable (33.3% pre-pilot vs. 34.9% post-pilot). The overall rate of stroke mimics was unchanged (66.7% vs. 66.5%). However, the distribution of mimics shifted: infection (11.2% to 8.5%), metabolic (5.4% to 3.2%), and Bell’s palsy (3.1% to 1.0%) decreased, while dizziness (12.0% to 14.4%) and functional presentations (1.6% to 4.1%) increased. Other mimic categories showed minimal change. No reduction in the proportion of true strokes was observed. Conclusions Empowering nurses to cancel EMS-activated stroke codes did not alter the overall yield of stroke activations but was associated with a reduction in easily recognized non-stroke presentations. Remaining mimics were predominantly conditions that are clinically difficult to distinguish from stroke. The findings support the safety and potential efficiency benefits of nurse-led stroke code triage. Conflict of interest Michael Teitcher: nothing to disclose